Abstract

In recent years, the use of artificial intelligence methodologies has demonstrated effectiveness in improving the safety and efficiency of building evacuations during emergencies. However, research in this field has been scattered, with limited understanding of current challenges and emerging trends. This paper addresses this gap by providing a systematic review of literature published between 2019 and 2023. This review enables researchers and practitioners to identify key areas of advancement and focus their efforts on innovative solutions for modeling building evacuation plans. Articles were categorized into groups based on inclusion and exclusion criteria, and relationships between them were analyzed. A total of 121 articles were included in the literature review. The structure of the document is divided into four parts: (a) description of the evacuation management phases in buildings, (b) classification of studies by machine learning techniques, (c) conceptualization and description of the application areas of artificial intelligence for modeling evacuation plans and (d) discussion of the main challenges and future directions. Finally, the phase: initial detection and response, the group: supervised learning, the subgroup: artificial intelligence applications for modeling and simulation of evacuations, and the application area: escape route efficiency evaluation makes the largest contributions.